This Accepted Manuscript has not been copyedited and formatted. The final version may differ from this version.
Research Article: New Research | Disorders of the Nervous System
Clonal analysis of newborn hippocampal dentate granule cell proliferation
and development in temporal lobe epilepsy
Clonal analysis of newborn granule cells in epilepsy
Shatrunjai P. Singh
1,2
1
1,4
, Candi L. LaSarge , Amen An
1
, John J. McAuliffe and Steve C. Danzer
1
Department of Anesthesia, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio 45229
2
Molecular and Developmental Biology Program, University of Cincinnati, Cincinnati, OH 45237
3
Departments of Anesthesia and Pediatrics, University of Cincinnati, Cincinnati, OH 45267
1,2,3
4
Department of Neuroscience, McMicken College of Arts and Sciences, University of Cincinnati, Cincinnati, OH
45221
DOI: 10.1523/ENEURO.0087-15.2015
Received: 12 August 2015
Revised: 23 November 2015
Accepted: 1 December 2015
Published: 24 December 2015
Author Contributions: S.P.S., C.L., and A.A. performed research; S.P.S., C.L., J.J.M., and S.C.D. analyzed
data; S.P.S., C.L., J.J.M., and S.C.D. wrote the paper; S.C.D. designed research.
Funding: NINDS: 2R01-NS-065020. Albert J. Ryan Fellowship Award; American Heart Association Pre-Doctoral
Award;
Conflict of Interest: Authors report no conflict of interest.
National Institute of Neurological Disorders and Stroke (SCD, 2R01-NS-065020). Albert J. Ryan Foundation
fellowship (SPS). American Heart Association Pre-Doctoral Award (SPS).
Correspondence should be addressed to: Dr. Steve C. Danzer, 3333 Burnet Avenue, ML 2001, Cincinnati,
Ohio 45229-3039, (513) 636-4526 (phone), (513) 636-7337 (fax). Email: [email protected]
Cite as: eNeuro 2015; 10.1523/ENEURO.0087-15.2015
Alerts: Sign up at eneuro.org/alerts to receive customized email alerts when the fully formatted version of this
article is published.
Accepted manuscripts are peer-reviewed but have not been through the copyediting, formatting, or proofreading
process.
This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International
(http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution and reproduction in any
medium provided that the original work is properly attributed.
Copyright © 2015 Society for Neuroscience
eNeuro
http://eneuro.msubmit.net
eN-NWR-0087-15R1
Clonal analysis of newborn hippocampal dentate granule cell proliferation
and development in temporal lobe epilepsy
This is a confidential document and must not be discussed with others, forwarded in any
form, or posted on websites without the express written consent of eNeuro.
Title Page
1
2
1. Manuscript Title (50 word maximum)
3
4
5
Clonal analysis of newborn hippocampal dentate granule cell proliferation and
development in temporal lobe epilepsy
6
2. Abbreviated Title (50 character maximum)
7
Clonal analysis of newborn granule cells in epilepsy
8
9
10
3. List all Author Names and Affiliations in order as they would appear in the published
article
11
12
13
14
15
16
17
18
19
Shatrunjai P. Singh1,2 , Candi L. LaSarge1, Amen An1,4 , John J. McAuliffe1 and Steve C.
Danzer1,2,3
20
21
22
4. Author Contributions: Each author must be identified with at least one of the following:
Designed research, Performed research, Contributed unpublished reagents/ analytic tools,
Analyzed data, Wrote the paper.
23
24
25
SPS performed research, analyzed data and wrote the paper. CLS performed research, analyzed
data, wrote paper. AA performed research. JJM analyzed data and wrote the paper. SCD
designed research, analyzed data and wrote the paper.
1
Department of Anesthesia, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio
45229; 2Molecular and Developmental Biology Program, University of Cincinnati, Cincinnati,
OH 45237; 3Departments of Anesthesia and Pediatrics, University of Cincinnati, Cincinnati, OH
45267; 4McMicken College of Arts and Sciences, Department of Neuroscience, University of
Cincinnati, Cincinnati, OH 45221
26
27
5. Correspondence should be addressed to (include email address)
28
29
30
31
32
33
Dr. Steve C. Danzer
3333 Burnet Avenue, ML 2001
Cincinnati, Ohio 45229-3039
(513) 636-4526 (phone)
(513) 636-7337 (fax)
Email: [email protected]
1
34
35
6. Number of Figures: 6
36
7. Number of Tables: 1
37
8. Number of Multimedia: 0
38
9. Number of words for Abstract: 239
39
10. Number of words for Significance Statement: 111
40
11. Number of words for Introduction: 430
41
12. Number of words for Discussion: 1895
42
13. Acknowledgements
43
44
45
46
47
48
49
50
51
This work was supported by the National Institute of Neurological Disorders and Stroke (SCD,
2R01-NS-065020), the Albert J. Ryan Foundation fellowship (SPS) and the American Heart
Association Pre-Doctoral Award (SPS). The content is solely the responsibility of the authors
and does not necessarily represent the official views of the National Institute of Neurological
Disorders and Stroke, the National Institutes of Health, or the Albert J. Ryan Foundation or the
American Heart Association. We would like to thank Matthew Kofron and Micheal Muntifering
(Confocal Imaging Core, Cincinnati Children’s Hospital, Cincinnati, OH) for technical advice on
imaging modalities used for this study. We would also like to thank Keri Kaeding and Mary
Dusing for useful comments on the earlier versions of this manuscript.
52
53
14. Conflict of Interest
54
Authors report no conflict of interest
55
56
15. Funding sources
57
National Institute of Neurological Disorders and Stroke (SCD, 2R01-NS-065020)
58
Albert J. Ryan Foundation fellowship (SPS)
59
American Heart Association Pre-Doctoral Award (SPS)
60
2
61
Abstract
Hippocampal dentate granule cells are among the few neuronal cell types generated
62
63
throughout adult life in mammals. In the normal brain, new granule cells are generated from
64
progenitors in the subgranular zone and integrate in a typical fashion. During the development of
65
epilepsy, granule cell integration is profoundly altered. The new cells migrate to ectopic
66
locations and develop misoriented “basal” dendrites. Although it has been established that these
67
abnormal cells are newly-generated, it is not known whether they arise ubiquitously throughout
68
the progenitor cell pool or are derived from a smaller number of “bad actor” progenitors. To
69
explore this question, we conducted a clonal analysis study in mice expressing the Brainbow
70
fluorescent protein reporter construct in dentate granule cell progenitors. Mice were examined
71
two months after pilocarpine-induced status epilepticus; a treatment that leads to the
72
development of epilepsy. Brain sections were rendered translucent so that entire hippocampi
73
could be reconstructed and all fluorescently-labeled cells identified. Our findings reveal that a
74
small number of progenitors produce the majority of ectopic cells following status epilepticus,
75
indicating that either the affected progenitors or their local micro-environments have become
76
pathological. By contrast, granule cells with “basal” dendrites were equally distributed among
77
clonal groups. This indicates that these progenitors can produce normal cells and suggests that
78
global factors sporadically disrupt the dendritic development of some new cells. Taken together,
79
these findings strongly predict that distinct mechanisms regulate different aspects of granule cell
80
pathology in epilepsy.
81
82
83
3
84
Significance Statement
85
Epileptogenic injuries disrupt adult neurogenesis, leading to the abnormal integration of adult-
86
generated granule cells. The newborn cells exhibit a variety of pathologies, including dendritic
87
abnormalities and migration defects. It was not known, however, whether all progenitors
88
contributed equally to the accumulation of these abnormal cells or whether a distinct subset of
89
progenitors was responsible. Here, we performed a clonal analysis study of progenitor cell
90
activity following status epilepticus. Our results reveal that a small subset of progenitors
91
produces the majority of ectopic granule cells, while cells with abnormal dendrites arise
92
ubiquitously throughout the progenitor pool. Together, these findings demonstrate a newly-
93
understood complexity among progenitors in producing abnormal granule cells in epilepsy.
94
4
95
Introduction
96
Hippocampal dentate granule cells (DGCs) are generated throughout life from progenitor
97
cells located in the subgranular zone, a proliferative region located between the granule cell body
98
layer and the hilus. A subset of progenitor cells in this region expresses the Gli1 transcription
99
factor, a Krüppel family zinc-finger protein activated by the sonic hedgehog-signal transduction
100
cascade (Ahn and Joyner, 2005, Palma et al., 2005). Sonic hedgehog is a key regulator of cell
101
proliferation (Lai et al., 2003). Gli1-expressing type-1 progenitor cells are morphologically
102
characterized by the presence of a radial process that projects into the dentate inner molecular
103
layer. They exhibit the stem cell characteristics of self-renewal and multipotency and give rise to
104
intermediate progenitors (type-2 cells; transient amplifying cells); which, in turn, give rise to
105
postnatally-generated dentate granule cells. Type-1 cells can also give rise to astrocytes
106
(Bonaguidi et al., 2012).
107
Adult-born DGCs are especially vulnerable to epileptogenic insults. Cells born in the
108
109
weeks before and after an insult develop morphological and functional abnormalities (Parent et
110
al., 2006, Jessberger et al., 2007, Walter et al., 2007, Murphy et al., 2011, Santos et al., 2011).
111
Following epileptogenic insults, adult-born DGCs populate the dentate hilus (hilar ectopic
112
granule cells), a region they rarely occupy in normal animals (Scharfman et al., 2000). Afferent
113
inputs to these ectopic DGCs are abnormal, and the cells can exhibit atypical bursting properties
114
(Zhan et al., 2010, Myers et al., 2013, Althaus et al., 2015). DGCs with basal dendrites are also
115
common in the epileptic brain, a feature typically absent from non-epileptic rodent DGCs. Basal
116
dendrites are hypothesized to form recurrent circuits and promote hyperexcitability within the
117
hippocampus (Ribak et al., 2000, Shapiro et al., 2008).
5
118
119
Although it is well established that abnormal DGCs are derived from adult-progenitor
120
cells, it is not known whether all progenitors contribute equally to the production of abnormal
121
cells, or whether distinct subsets of progenitors preferentially produce them. Answering this
122
question will provide novel insights into the mechanisms underlying aberrant granule cell
123
accumulation. Equal participation among progenitors suggests systemic changes in the factors
124
regulating granule cell development, while unequal participation suggests regional disruption of
125
neurogenic niches or intrinsic changes within individual progenitors. Here, we utilized a
126
conditional Brainbow reporter line driven by an inducible Gli1-CreERT2 promotor construct to
127
lineage-trace clones arising from Gli1-expressing granule cell progenitors in the pilocarpine
128
model of epilepsy. Brains were rendered translucent using a novel clearing agent. Hippocampi
129
were imaged in their entirety to identify and characterize groups of daughter cells, known as
130
“clonal clusters,” each of which originates from a single, labeled progenitor.
131
132
133
134
135
6
136
Methods
137
Animals
All methods involving animals were approved by the Institutional Animal Care and Use
138
139
Committee of the Cincinnati Children's Hospital Research Foundation and conform to NIH
140
guidelines for the care and use of animals. For the present study, hemizygous Gli1-CreERT2 mice
141
(Ahn and Joyner, 2005) (https://www.jax.org/strain/007913) were crossed to mice homozygous
142
for a Gt(ROSA)26Sortm1(CAG-Brainbow2.1)Cle/J “Brainbow” reporter construct (Cai et al., 2013)
143
(https://www.jax.org/strain/013731) to generate double transgenic Gli1-CreERT2::Brainbow
144
mice. All animals were on a C57BL/6 background. A total of 30 double-transgenic mice were
145
randomly assigned to control or treatment (pilocarpine-induced status epilepticus) groups for the
146
present study. Postnatal tamoxifen treatment of these mice restricts CreERT2 expression to type-1
147
cells in the hippocampal subgranular zone (Ahn and Joyner, 2005, Murphy et al., 2011, Hester
148
and Danzer, 2013). Tamoxifen-induced activation of Cre-recombinase causes random excision
149
and/or inversion between multiple pairs of lox sites, leading to the expression of one of four
150
possible different fluorescent proteins in progenitor cells and all their progeny (Livet et al.,
151
2007). To facilitate morphological analyses, only cells expressing the cytoplasmic red or yellow
152
fluorescent proteins (RFP or YFP) were examined. Cells expressing cyan fluorescent protein
153
(CFP) were excluded because morphological details were poorly revealed by this membrane-
154
bound protein. GFP-expressing cells were not observed in any of the animals, consistent with
155
prior work (Calzolari et al., 2015).
156
157
Tamoxifen-induced cell labelling and pilocarpine treatment
To achieve sparse labelling of progenitor cells, mice were given three injections of
158
159
tamoxifen (250 mg/kg, s.c.) on alternate days during postnatal week seven (Fig. 1A). At eight
7
160
weeks of age, all mice received methyl scopolamine nitrate in sterile saline (1 mg/kg, s.c.),
161
followed fifteen minutes later by either pilocarpine (420 mg/kg, s.c.; n=25) or saline solution
162
(controls, n=5). Animals were monitored behaviorally for seizures and the onset of status
163
epilepticus (defined as continuous tonic-clonic seizures). Following three hours of status
164
epilepticus (SE) mice were given two injections of diazepam, ten minutes apart (10 mg/kg, s.c.),
165
to alleviate seizure activity. Mice were given sterile Ringers solution as needed to restore
166
pretreatment body weight and were then returned to their home cages, where they were provided
167
with food and water ad libitum. Mice were housed under a 14/10 hour light/dark cycle to
168
optimize breeding (Fox et al., 2007). Of the 25 mice randomly assigned to pilocarpine treatment,
169
12 (48%) survived to the end of the experiment, yielding a final group consisting of 7 males and
170
5 females. There was no mortality among the five control animals (3 males and 2 females).
171
172
Tissue preparation for whole hippocampal imaging
At 16 weeks of age, animals were euthanized by intraperitoneal injection of 100 mg/kg
173
174
pentobarbital. The mice were perfused through the ascending aorta with ice-cold phosphate-
175
buffered saline (0.1M PBS) containing 1 U/ml heparin for 30 seconds at 10 ml/min, immediately
176
followed by a 2.5% paraformaldehyde + 4% sucrose solution in 0.1M PBS at 25oC for ten
177
minutes. Brains were removed, dissected into left and right hemispheres and post-fixed in the
178
same solution overnight at 4°C. Brain hemispheres were cryoprotected in 10%, 20% and 30%
179
sucrose in PBS for 24, 24 and 48 hours, respectively. The hemispheres were then frozen in
180
isopentane cooled to −23°C with dry ice and stored at −80°C until sectioning. Brain hemispheres
181
were thawed in PBS for scale clearing and 300 µm coronal sections were cut on a tissue slicer
182
(Campden Instruments, Lafayette, USA). Sections were transferred to 24 multiwell tissue culture
8
183
plates (Becton Dickinson, New Jersey, USA), maintaining their septo-temporal order. Sections
184
were incubated for optical clearing in ScaleA2 for two weeks at 4°C (Hama et al., 2011). At least
185
one hemisphere from each animal was used for clonal analysis, and in three cases both
186
hemispheres were used (2 SE and 1 control). No significant differences in cluster composition
187
were found between hemispheres within animals (Mann-Whitney rank sum test), so hemispheres
188
were pooled by animal for statistical analysis. The remaining hemispheres were used for
189
immunohistochemical characterization of Brainbow-labeled cells.
190
191
Tissue preparation for immunohistochemistry
Unused hemispheres from a subset of animals (n=5) were sectioned coronally on a
192
193
cryostat at 60 μm and mounted to gelatin-coated slides. Sections were immunostained with
194
mouse anti-nestin (1:100; Millipore), chicken anti-GFAP (1:500; Chemicon), goat anti-
195
doublecortin (1:250; Santa Cruz), mouse anti-calretinin (1:200; Millipore) or guinea pig anti-
196
calbindin-D-28K (1:200; Sigma). Alexa Fluor 405 goat anti-mouse, 488 goat anti-chicken, 594
197
goat anti-mouse, 647 donkey anti-goat or Alexa Fluor 647 goat anti-guinea pig secondary
198
antibodies were used (Invitrogen). Tissue was dehydrated in alcohol series and cleared in
199
xylenes, and coverslips were secured with mounting media (Krystalon; Harleco).
200
201
Confocal Microscopy
ScaleA2 cleared hippocampal sections were imaged on a Nikon A1R GasAsP confocal
202
203
system attached to a motorized Nikon Eclipse Ti inverted microscope (Nikon Instruments, New
204
York, USA). This system was used to capture three-dimensional image stacks through the z-
205
depth of the tissue at 1 μm steps using a 10X Plan Apo λ objective (NA=0.25) at 1X optical
9
206
zoom (field size 1024 x 1024 pixels, 1.23 pixels/µm). These 10X image stacks were used to
207
identify clonal clusters, defined here as cells expressing the same fluorophore and contained
208
within a 150 μm radius of the clone center (Bonaguidi et al., 2011, Calzolari et al., 2015).
209
Identified clonal clusters were then imaged using a 40X Plan Apo IR DIC- Water Immersion
210
objective (NA=1.3) at 1X optical zoom (field size 1024 x 1024 pixels, 0.31 pixels/um). All cells
211
selected for analysis were brightly labeled with RFP or YFP and had their somas fully contained
212
within the tissue section. The investigator was blind to treatment group during all image
213
collection and data analysis.
214
215
3-dimensional hippocampal reconstructions
Confocal z-series image stacks were converted into 8 bit RGB tiff files. Reconstruct
216
217
Software (John C. Fiala, the National Institutes of Health) (Lu et al., 2009) was used to septo-
218
temporally align sections (10X images) for each hippocampus (Fig. 1B). Aligned z-stacks were
219
imported into Neurolucida software for analysis (Version 11.01, Microbrightfield Inc.,Williston,
220
VT). Borders of the granule cell body layer were traced at z-intervals of 100 μm to recreate the
221
whole hippocampus.
222
223
Morphological classification
Higher magnification images (40X) were used to categorize cells within each cluster as
224
225
follows: 1) Type-1 cell, with a small cell body located in the subgranular zone and a single,
226
radial process that projects through the granule cell layer and terminates in the inner molecular
227
layer. Type-1 progenitor cells express nestin and glial fibrillary acidic protein (GFAP). 2) Type-
228
2/3 cells, with a cell body located in the subgranular zone and short, aspiny processes projecting
10
229
parallel to the plane of the granule cell body layer. Since their appearance is morphologically
230
similar, we did not attempt to distuingish between type-2 and 3 cells, or the different subtypes of
231
type-2 cells. Type-2 and 3 cells express the cellular proliferation marker doublecortin. 3)
232
Immature granule cells, with a cell body in the granule cell body layer and aspiny dendrites that
233
project radially through the granule cell body layer, but terminate prior to reaching the
234
hippocampal fissure (typically with growth cones at the tips). These cells occasionally possessed
235
short, aspiny basal dendrites and express calretenin. 4) Normal mature granule cells, with their
236
somas located in the granule cell body layer and spine-coated dendrites projecting to the
237
hippocampal fissure. Mature granule cells express calbindin. 5) Hilar ectopic granule cells, with
238
spiny dendrites and a cell body located in the hilus (at least two cell body distances, ≈20 μm,
239
away from the granule cell layer-hilar border). 6) Mature granule cells with basal dendrites,
240
possessing all the features of normal mature granule cells (see 4), but with at least one dendrite
241
originating from the hilar side of the soma (i.e. arising from a region below the soma midline).
242
Cells with basal dendrites projecting into either the dentate hilus, or the dentate molecular layer
243
(recurrent basal dendrites), were included in this measure. Only granule cells with clearly visible
244
axons were scored for basal dendrites. Basal dendrites are often thin and difficult to visualize in
245
deeper regions of the tissue. Well-developed axons should be present on all mature granule cells.
246
Confirming that these axons can be visualized limits the entry of false negatives into the data set.
247
If the axon can be visualized, then any basal dendrites, which are typically of higher caliber than
248
the axon, should also be detectable. 7) Astrocytes, defined as cells with a small cell body, located
249
anywhere within the dentate gyrus and possessing numerous thin, aspiny process projecting
250
outwards in a stellate fashion. For review of granule cell developmental markers see
251
(Kempermann et al., 2004, Bonaguidi et al., 2012, Kempermann et al., 2015).
11
252
Statistical Analysis
Microsoft SQL Server (version 2012) was used to query the dataset for different clone
253
254
compositions, and statistical analysis was performed using R (Version 0.98.109) or Sigma Plot
255
(version 12.5). Sex and treatment effects were determined using two-way ANOVA. Individual
256
group differences were determined using the Holm-Sidak method for all ANOVA results.
257
Parametric tests were used for data that met assumptions for normality and equal variance. Data
258
that failed assumptions of normality and equal variance were either transformed as noted in the
259
text to meet these assumptions, or were analyzed using non-parametric alternative tests. Actual
260
tests used are noted in the text. Values presented are means ± SEM (least square means for two-
261
way ANOVA data) or medians [range], as appropriate. Details of statistical tests are given in
262
Table 1.
263
The statistical analysis for the frequency/distribution of ectopic cells and basal dendrites
264
265
was performed using the binomial distribution (to compute probabilities of combinatorial
266
events). The experiment-wise error was conservatively set at 0.001 (Cumming, 2010).
267
Corrections for multiple comparisons were done using a Bonferroni correction. For clusters
268
containing ectopic cells, the resultant p-value for significance for the pilocarpine treated animals
269
was calculated to be 4.17x10-6 (0.001/240). Similarly, for clusters containing DGCs with basal
270
dendrites, the probability of a single trial success was 0.0614 and the critical p-value was
271
calculated to be 5.41x10-6 (0.001/185).
272
273
Figure Preparation
12
Maximum projections from z-series stacks were prepared using NIS-Elements Ar
274
275
Microscope Imaging Software (version 4.0). Contrast, brightness, montage adjustments and
276
figure preparation were done using Adobe Photoshop CS5 (version 12.0). Identical filtering and
277
adjustments to brightness and contrast were done for images meant for comparison. Tableau
278
(version 8.0) and Microsoft Excel (version 2013) were used to create graphs and visualizations.
279
The image in figure 4 was cropped to remove neuronal structures above and below the cell of
280
interest that would otherwise obscure it (Walter et al., 2007, McAuliffe et al., 2011, Murphy et
281
al., 2012). This image is best viewed as a neuronal reconstruction, similar to traditional
282
neuroanatomical techniques (Danzer et al., 1998), rather than a standard photomicrograph.
283
284
285
286
13
287
Results
288
In vivo lineage tracing of individual Gli1-expressing progenitor cells in the adult mouse
289
hippocampus
290
To study the proliferative activity of a cohort of Gli1-expressing granule cell progenitors
291
in epilepsy, we treated double transgenic Gli1-CreERT2::Brainbow reporter mice with tamoxifen
292
at post-natal week seven to lineage-trace these cells. A small cohort of animals was perfused two
293
days later, revealing an average of two type-1 cells per 300 μm hippocampal coronal section
294
(Fig.1), an optimal labeling sparsity for identifying individual clones (Bonaguidi et al., 2011).
295
Gli1 expression has been shown to mark multipotent type-1 stem cells, which give rise to type-
296
2/3 stem cells and other differentiated progeny (Encinas et al., 2011). Animals in the main study
297
groups received either saline or pilocarpine one week after tamoxifen treatment (Fig.1A).
298
Pilocarpine induces acute status epilepticus (SE) and the subsequent development of epilepsy a
299
few weeks later (Turski et al., 1983). Animals were killed two months after pilocarpine treatment
300
– when spontaneous seizures are typically frequent (Castro et al., 2012, Hester and Danzer,
301
2013). Hippocampi were imaged in their entirety to identify all fluorescently-labeled cells
302
(Fig.1B). Brainbow fluorophore expression was strictly tamoxifen dependent and was limited to
303
dentate granule cells, astrocytes and sub-granular zone progenitor cells (Fig.1C). We first
304
assessed whether status epilepticus altered the number of clonal clusters per hippocampus
305
(control, female, n=2 mice, 4.0±5.8 clusters per hippocampus; control, male, n=3, 13.3±4.8; SE,
306
female, n=5, 24.9±4.2; SE, male, n=7, 11.0±2.9). The effect of status epilepticus was found to be
307
dependent on animal sex (Fig.2; p=0.025, two-way ANOVA). Post-hoc tests revealed a
308
significant increase in female, but not male, mice in clusters per hippocampus relative to controls
309
(p = 0.012, Holm-Sidak method) and significantly more clusters in females vs. males within the
14
310
pilocarpine-treated groups (p = 0.017, Holm-Sidak method). Differences between sexes could
311
reflect differential numbers of progenitors prior to pilocarpine treatment and differential behavior
312
of progenitors after treatment. Greater apoptosis of quiescent progenitor cells or entire clonal
313
groups in males, for example, would reduce the number of clonal clusters. Sexually dimorphic
314
changes in neurogenesis have been observed following early-life stress in rodents (Loi et al.,
315
2014). The present findings suggest that dimorphic responses to adult status epilepticus also
316
occur.
317
318
Status epilepticus increases the average number of cells per clonal cluster
Increased hippocampal neurogenesis and cell survival have been consistently
319
320
demonstrated in epilepsy models (Bengzon et al., 1997, Parent et al., 1997, Gray and Sundstrom,
321
1998, Parent et al., 1998). In addition to an increase in the number of clonal clusters in female
322
mice, the present work also revealed an increase in the mean size of individual clones.
323
Specifically, mean clonal size increased from 3.0±0.7 cells/cluster in controls to 5.1±0.5 in SE
324
mice (Fig.1, p=0.033, SE (n=12) vs. control (n=5), two-way ANOVA). In contrast to cluster
325
number, however, no differences between males and females were found for cluster size
326
(p=0.822) nor was there an interaction between treatment and sex (p=0.107). Similarly, no
327
additional sex differences or interactions were found in pre-tests for all additional data presented
328
here (data not shown), so males and females were binned for all further statistical analyses.
329
Taken together, these data suggest that the increase in new granule cells after SE is likely due to
330
increased proliferation among individual progenitors (increased cluster size) and/or reduced
331
apoptosis of their progeny (more clusters, increased cluster size).
332
15
333
Status epilepticus promotes terminal differentiation of hippocampal progenitor cells
While neurogenesis is increased after an acute epileptogenic injury, it can decrease in
334
335
chronic epilepsy (Hattiangady et al., 2004, Danzer, 2012). A growing body of literature
336
demonstrates that progenitor cell pools can be depleted as progenitors proceed through multiple
337
rounds of division, ultimately leading to terminal differentiation (Ledergerber et al., 2006,
338
Encinas et al., 2011). Reduced neurogenesis in chronic epilepsy, therefore, could be a direct
339
consequence of increased progenitor cell activity early in the disease.
340
To assess whether this might be the case, we used morphological criteria to classify cells
341
342
as type-1 progenitors, type-2/3 progenitors, immature granule cells, mature granule cells or
343
astrocytes (Fig. 2A). The accuracy of this morphological classification was confirmed by
344
immunocharacterization of the different cell types quantified (Fig. 2B). This lineage analysis
345
revealed a significant reduction in the progenitor cell pool in status-exposed animals. Compared
346
to control animals, status animals exhibited an 84% reduction in type-1 cells (p=0.013, SE
347
(n=12) vs. control (n=5), Mann-Whitney Rank Sum Test). No difference in the percentage of
348
type-2/3 cells was found (p=0.870, SE (n=12) vs. control (n=5), Mann-Whitney RST). Overall
349
there was a 58% reduction in the percentage of clusters containing either type-1 or type-2/3
350
progenitors, dropping from 46.2±8.1% in control animals to 18.6±5.0% in animals exposed to
351
status (Fig. 2E; p=0.009, SE (n=12) vs. control (n=5), Two-tailed t-test). Symmetric self-
352
renewing clusters (composed of two type-1 cells) decreased from a median of 4.5% [range 0-
353
16.7%] of clusters/animal in controls to nil in SE animals (p=0.006, SE (n=12) vs. control (n=5),
354
Mann-Whitney RST). The significant shift away from progenitors in animals exposed to status
355
was mirrored by a non-significant trend in the proportion of mature granule cells produced
16
356
(p=0.065, SE (n=12) vs. control (n=5), Mann-Whitney RST) and a significant increase in the
357
percentage of clusters that were “fully differentiated,” meaning the clusters contained only
358
granule cells and astrocytes, and were devoid of progenitors (p=0.010, SE (n=12) vs. control
359
(n=5), Two-tailed t-test). No differences in the proportions of immature granule cells (p=0.299,
360
Mann-Whitney RST) or astrocytes (p=0.223, Mann-Whitney RST) were found.
361
362
Ectopic DGCs appear in clonal clusters in which majority of the cells are ectopic
363
Hilar ectopic granule cells are a common pathology seen in temporal lobe epilepsy. These
364
neurons are newly-generated, arising after an epileptogenic brain insult, and are implicated in the
365
development of epilepsy (Hester and Danzer, 2013). The percentage of newborn cells found in
366
the hilar region of status-exposed animals was significantly increased relative to controls (Fig.3;
367
p=0.004; SE, 63 of 1238 DGCs (5.09%), Control, 1 of 192 cells (0.52%); Mann-Whitney RST),
368
consistent with previous studies (Parent et al., 2006). Clonal analysis revealed that ectopic cells
369
were concentrated within a small number of clusters. Specifically, the 63 identified ectopic cells
370
from animals exposed to status were contained within only 14 clusters (Fig. 3B). Within these
371
clusters, 76.8% of all cells were ectopic (Fig. 3B).
372
The probability of finding a set number of ectopic cells, S, in a cluster containing a total
373
374
of N cells was computed for all values of S from 1 to N (e.g. when S=N, 100% of the cells in the
375
cluster are ectopic). The minimum number of ectopic cells S, at which the p-value reaches the
376
target p-value for significance (p=4.17x10-6; see methods for calculations) was determined for
377
every cluster and compared to the observed number of ectopic cells. We found that six of the 14
378
clusters with ectopic cells exceeded the threshold p-value for significance (binomial p-value <
17
379
4.17x10-6; Fig3). The binomial probability of finding six of 240 clusters exceeding this value is
380
statistically minute (less than 1.33x10-21). Finding one cluster that exceeded the value would
381
indicate that ectopic cells are not randomly distributed, and in the present study six such clusters
382
were observed. These results provide overwhelming evidence that the accumulation of ectopic
383
cells in certain clusters is not a random event.
384
385
DGCs with basal dendrites occur in clonal clusters in which majority of the cells do not have a
386
basal dendrite
Another common pathology observed in the epileptic brain is the presence of dentate
387
388
granule cells with basal dendrites (Spigelman et al., 1998, Buckmaster and Dudek, 1999). In the
389
current study, 6.14% of granule cells from status-exposed animals possessed basal dendrites. By
390
contrast, only one cell with a basal dendrites was found among the control animals (p=0.034; SE,
391
43 of 700 DGCs, Control, 1 of 101 cells; Mann-Whitney RST). The 43 basal dendrite-possessing
392
granule cells from SE mice were distributed among 31 of 185 clonal clusters (Fig.4). Among
393
these 31 clusters that contained a cell with a basal dendrite, 23 had only a single basal dendrite-
394
possessing cell; six had two; one had three and one had five cells with basal dendrites (Fig.4).
395
Using the significance criterion described in the methods (similar to that used for ectopic cells),
396
there were no clusters that contained a significant number of DGCs with basal dendrites (Fig.4).
397
Indeed, even the biggest cluster - containing 17 DGCs with five harboring a basal dendrite -
398
failed to reach significance even if the experiment-wise alpha is relaxed to 0.05 (p= 0.054;
399
threshold p value for α=0.001, 5.41x10-6; for α=0.05, 2.60x10-4).
400
401
Clonal clusters have more type-1 cells in dorsal hippocampus
18
Previous studies have reported topographical differences within the hippocampus with
402
403
respect to neurogenesis, cell densities, functional properties and electrophysiological properties
404
(Jinno, 2011, Kheirbek and Hen, 2011; Jhaveri et al., 2015). To determine whether there are
405
dorsal-ventral differences in progenitor cell behavior, we correlated cluster composition with
406
cluster bregma level (Paxinos and Franklin, 2001). Within SE animals, the number of type-1
407
cells was significantly correlated with bregma level, with greater numbers of cells identified in
408
more dorsal regions (Fig.5; p=0.0037; Spearman Rank Order Correlation). This effect persisted
409
when bregma levels were correlated with the proportion of type-1 cells at each level (type-1
410
cells/total cells), suggesting that differences in total cell numbers cannot account for the finding
411
(p=0.0238; Spearman Rank Order). The numbers and proportions of mature, immature, type 1,
412
type 2, astrocytic, ectopic and basal dendrite-possessing cells were not significantly correlated
413
with bregma level (data not shown). No significant correlations were found between bregma
414
level and the number or proportion of any cell types in control animals (data not shown).
415
416
417
418
419
19
420
Discussion
Abnormal hippocampal granule cells are common in animal models of temporal lobe
421
422
epilepsy (Rolando and Taylor, 2014) and in tissue from patients with the disease (Sutula et al.,
423
1988, Parent et al., 2006). Prior studies have established that many of these abnormal cells are
424
adult-generated (Walter et al., 2007, Kron et al., 2010). In the present study, we queried whether
425
two important abnormalities – ectopic DGCs and DGCs with basal dendrites – are derived with
426
equal likelihood from the entire progenitor pool, or whether they are preferentially produced by a
427
subset of progenitors. We found that ectopic granule cells were highly concentrated within
428
distinct clonal clusters: many containing only ectopic cells. By contrast, cells with basal
429
dendrites were relatively evenly distributed among clones. These findings strongly suggest the
430
existence of distinct mechanisms regulating ectopic cell migration and basal dendrite formation.
431
A second key finding provides new insights into the bimodal changes in neurogenesis rates
432
observed in epileptic animals. Neurogenesis increases acutely following an epileptogenic insult;
433
however, animals with chronic epilepsy exhibit reduced neurogenesis. Depletion of the
434
progenitor pool - potentially as a direct consequence of the early increase in neurogenesis - has
435
been hypothesized to account for these changes (Hattiangady et al., 2004). Our data provide
436
direct evidence that this is occurring, with a 70% increase in the number of daughter cells/clone
437
and a corresponding decrease in the percentage of actively dividing and self-renewing clones
438
(Fig.6).
439
440
Limitations of the current study
441
The present study uses clonal analysis methodology previously validated for the dentate
442
gyrus (Bonaguidi et al., 2011). Nonetheless, it is expected that progenitor cells labeled with the
20
443
same fluorophore will occasionally appear in close proximity to one another, leading to the false
444
conclusion that they represent a single clonal cluster. Conclusions based on rare clonal events,
445
therefore, should be made with caution if the observation could also be accounted for by a false
446
merging of clusters. As an example, it is possible that some of the clones with mixes of ectopic
447
and normally positioned cells shown in figure 3 are actually merged clonal clusters. Notably,
448
however, our findings in control animals are remarkably similar to Bonaguidi and colleagues
449
(2011) work using the Nestin-CreERT2 driver line to label progenitor cells. We found a similar
450
distribution of cluster sizes (Fig.1D) in the animals and were able to reproduce key findings -
451
such as the occurrence of symmetric cell divisions - yielding two type-1 cells. One difference we
452
noted was a greater degree of neurogenesis among clusters using the Gli1 driver relative to the
453
nestin driver used previously. The nestin driver produced a roughly equal ratio of neurons to
454
astrocytes (Bonaguidi et al., 2011, Song et al., 2012); whereas neurons were much more common
455
in the present work (Fig.2D). The higher ratio of neurons to astrocytes is consistent with studies
456
using cell birthdating and viral-labeling approaches (Steiner et al., 2004), and may reflect
457
differences between Gli1 and nestin-expressing stem cells.
458
A second notable caveat is that the present study examined only progenitor cells labeled
459
460
in the week before status epilepticus, in order to examine the impact of status epilepticus on
461
progenitor cells (rather than their offspring, as would be achieved with earlier labeling), and to
462
ensure equivalent labeling of progenitor populations between control and epileptic animals. The
463
progenitor pool changes after status epilepticus, so labeling after the insult will presumably label
464
a population of progenitors that diverges from controls. Aberrant granule cell integration occurs
465
over a protracted time course, and includes immature cells born weeks before status, as well as
21
466
cells born months later. Whether the current findings will extend to populations generated at
467
other time points remains to be determined.
468
469
Clonal analysis of adult SGZ neurogenesis following status epilepticus
In the present study we employed an in vivo, genetic, sparse-labeling approach to mark
470
471
stem cells for lineage-tracing. This approach has been used previously to study neural stem cell
472
behavior in the sub-ventricular (Calzolari et al., 2015) and sub-granular (Suh et al., 2007,
473
Bonaguidi et al., 2011) proliferative zones in healthy animals. We combined this approach with
474
recently developed tissue clearing protocols, allowing us to generate three-dimensional
475
reconstructions of the entire rodent hippocampus; and the Brainbow reporter line, allowing us to
476
separate clonal groups by fluorochrome expression. Poor recombination with the Brainbow
477
reporter in CNS tissue limited our study to two colors, rather than the potential seven colors
478
evident in other tissues (Livet et al., 2007, Cai et al., 2013). Even with two colors, however, the
479
strategy provided sufficient spatial resolution for the study.
480
An additional advantage of the genetic approach is that it avoids disturbing the target
481
482
tissue with direct brain injections, as is needed for retroviral strategies (Hope and Bhatia, 2011,
483
Ming and Song, 2011). Retrovirus also targets Sox2+ type-2 progenitor cells (Suh et al., 2007),
484
while the Gli1 driver used here targets the parent type-1 cells (Ahn and Joyner, 2005) , so the
485
different strategies provide complimentary data.
486
487
488
22
489
Localized regulation of ectopic granule cell formation
Ectopic granule cells have been observed in a number of different epilepsy models. They
490
491
are hyper-excitable (Scharfman et al., 2000, Althaus et al., 2015), have atypical connections
492
within the hippocampal network (Scharfman and Pierce, 2012), and their numbers correlate with
493
the severity and duration of seizures (Hester and Danzer, 2013). The mechanisms underlying
494
ectopic cell migration, however, are unknown. A mechanistic understanding would provide new
495
insights into the development of therapeutic strategies for epilepsy. A putative mechanism that
496
could account for the current results is mislocation of progenitor cells from sub-granule zone to
497
the hilus during epileptogenesis (Parent et al., 2006). Alternatively, epileptogenic stimuli could
498
activate “dormant” progenitors trapped in the hilus during development (Gaarskjaer and
499
Laurberg, 1983, Scharfman et al., 2007). If one further presumes that daughter cells produced by
500
hilar progenitors would not have access to the necessary cues directing them to migrate into the
501
granule cell layer, the presence of entirely ectopic clonal groups could be accounted for.
502
Alternatively, seizures might lead to the localized disruption of migratory cues, like reelin
503
(Teixeira et al., 2012). Progenitor cells active in regions with disrupted cues would produce
504
daughter cells that fail to migrate correctly, while progenitors in regions with intact cues would
505
produce normal offspring. In support of this possibility, Parent and colleagues observed trains of
506
cells migrating on glial scaffolds into the hilus after seizures (Parent et al., 2006), suggesting that
507
localized changes in non-neuronal cells might play a role. Additional studies will be needed to
508
distinguish among these possibilities.
509
The finding that clonal groups with ectopic cells tend to be made up of entirely ectopic
510
511
cells is consistent with a Markov chain model (Lange, 2003). In a Markov model, there are two
23
512
states a progenitor cell can assume: Progenitors in state one give rise to DGCs correctly located
513
in the cell body layer; whereas progenitors in state two give rise to ectopic DGCs. For the model,
514
we assumed that at every mitotic cycle, cells could either stay in the same state, or transition
515
between states. The transition matrix specifies the probabilities of these transitions. Our data
516
show that the probability of a progenitor switching states is very low, and the probability that a
517
progenitor will remain in the same state is close to 100%, implying that cells that begin
518
producing ectopic cells will continue to do so, and cells that initially produce normal cells also
519
will continue to do so. Transitions between states appear to be very rare. Only 5 of 240 clusters
520
from SE mice contained a mixture of ectopic and correctly-located cells.
521
522
Temporal/global regulation of basal dendrite formation
Basal dendrites were distributed close to the predicted ratio among clonal clusters, and
523
524
tended to be present in clusters in which the majority of cells lacked this feature. Progenitors that
525
produce DGCs with basal dendrites, therefore, mostly produce morphologically normal DGCs.
526
This observation suggests a mechanism that could impact the development of daughter cells
527
from any progenitor, while also leaving most daughter cells unaffected. Such a mechanism might
528
affect the entire hippocampus, but only some of the time. Seizure activity clearly meets these
529
criteria, as seizures are episodic in nature, and when these seizures generalize, as is typical for
530
the pilocarpine model, the entire hippocampus will be affected. Recent work by Botterill and
531
colleagues (2015) supports this idea, with the demonstration that limbic kindling disrupts granule
532
cell integration more severely than kindling of non-limbic regions.
533
24
534
The idea that seizure activity might drive basal dendrite formation is supported by the
535
work of Nakahara and colleagues (Nakahara et al., 2009). They demonstrated that increasing
536
neuronal activity in hippocampal slice cultures stabilized the normally transient basal dendrites
537
typically present on immature granule cells. Under low activity conditions, developing granule
538
cells briefly possess basal dendrites one to two weeks after their birth, but subsequently reabsorb
539
these processes as they mature. Increasing neuronal activity, however, allowed these processes to
540
persist through cell maturity, perhaps through a neurotrophin dependent mechanism (Danzer et
541
al., 2002, Botterill et al., 2015). Whether a similar process occurs in vivo remains to be
542
determined, but the present findings are consistent with the idea that episodic increases in
543
activity (including seizures) might similarly stabilize granule cell basal dendrites – but only
544
among granule cells that happen to be at this particular developmental stage at the time of the
545
event.
546
547
Depletion of the granule cell progenitor pool
In the current study we found a decrease in progenitor cell numbers and an increase in
548
549
differentiated cells. A chronic decrease in neurogenesis has been observed previously in epileptic
550
animals (Hattiangady et al., 2004). Reduced neurogenesis could be the result of increased
551
progenitor cell quiescence, loss of functional progenitor cell division, decreased survival of
552
daughter cells, or an overall loss of progenitors. Our results provide evidence for the division-
553
coupled loss of type-1 progenitor cells as a key contributor to the chronic decline in
554
neurogenesis. We observed a decrease in quiescent and actively self-renewing progenitors, but
555
an increase in mature granule cells within clonal groups in animals following status. Activation
556
and terminal differentiation of quiescent progenitors would account for these observations.
25
557
Indeed, Encinas and colleagues (Encinas et al., 2011) observed a similar loss of stem cells during
558
the normal aging process in the mouse hippocampus. Using a genetic label they showed that
559
type-1 cells act as “disposable stem cells”: Once activated they tend to terminally differentiate
560
rather than continuing to cycle. Therefore, epileptic stimuli might accelerate the age-related loss
561
of progenitor cells from the dentate. Sierra and colleagues observed a similar reduction in the
562
progenitor cells pool following intrahippocampal injection of the convulsant kainic acid. In
563
contrast to the present results, however, they observed terminal differentiation of type-1 cells
564
into astrocytes (Sierra et al., 2015; see also Hattiangady and Shetty, 2010), rather than mature
565
granule cells as described here. This difference likely reflects the very different pathological
566
responses - and impacts on neurogenesis - between the two epilepsy models (Murphy et al.,
567
2012).
568
569
Concluding Remarks
570
Our results strongly suggest different mechanistic origins for ectopic DGCs and DGCs
571
with basal dendrites. Ectopic DGCs are highly localized to specific clonal clusters, implicating
572
the parent progenitor cell or the neurogenic niche in which the progenitor resides. By contrast,
573
basal dendrites appeared to be randomly distributed among clones, suggesting transient changes
574
acting throughout the hippocampus drive this pathology. Both abnormalities are implicated in the
575
development of epilepsy and associated comorbidities, and separate therapeutic strategies will
576
likely be required to mitigate these different abnormalities.
577
578
26
Reference
579
580
581
582
Ahn S, Joyner AL (2005) In vivo analysis of quiescent adult neural stem cells responding to
Sonic hedgehog. Nature 437:894-897.
583
584
585
Althaus AL, Sagher O, Parent JM, Murphy GG (2015) Intrinsic neurophysiological properties of
hilar ectopic and normotopic dentate granule cells in human temporal lobe epilepsy and a rat
model. J Neurophysiol. 113(4):1184-94.
586
587
588
Bengzon J, Kokaia Z, Elmer E, Nanobashvili A, Kokaia M, Lindvall O (1997) Apoptosis and
proliferation of dentate gyrus neurons after single and intermittent limbic seizures. Proc Natl
Acad Sci U S A 94:10432-10437.
589
590
Bonaguidi MA, Song J, Ming GL, Song H (2012) A unifying hypothesis on mammalian neural
stem cell properties in the adult hippocampus. Current opinion in neurobiology 22:754-761.
591
592
593
Bonaguidi MA, Wheeler MA, Shapiro JS, Stadel RP, Sun GJ, Ming GL, Song H (2011) In vivo
clonal analysis reveals self-renewing and multipotent adult neural stem cell characteristics. Cell
145:1142-1155.
594
595
596
Botterill JJ, Brymer KJ, Caruncho HJ, Kalynchuk LE (2015) Aberrant hippocampal neurogenesis
after limbic kindling: Relationship to BDNF and hippocampal-dependent memory. Epilepsy
Behav. 47:83-92.
597
598
Buckmaster PS, Dudek FE (1999) In vivo intracellular analysis of granule cell axon
reorganization in epileptic rats. Journal of neurophysiology 81:712-721.
599
600
Cai D, Cohen KB, Luo T, Lichtman JW, Sanes JR (2013) Improved tools for the Brainbow
toolbox. Nature methods 10:540-547.
601
602
603
Calzolari F, Michel J, Baumgart EV, Theis F, Gotz M, Ninkovic J (2015) Fast clonal expansion
and limited neural stem cell self-renewal in the adult subependymal zone. Nat Neurosci 18:490492.
604
605
606
Castro OW, Santos VR, Pun RY, McKlveen JM, Batie M, Holland KD, Gardner M, GarciaCairasco N, Herman JP, Danzer SC (2012) Impact of corticosterone treatment on spontaneous
seizure frequency and epileptiform activity in mice with chronic epilepsy. PLoS One 7:e46044.
27
607
608
609
Cumming G (2010) Replication, p rep, and confidence intervals: comment prompted by Iverson,
Wagenmakers, and Lee (2010); Lecoutre, Lecoutre, and Poitevineau (2010); and Maraun and
Gabriel (2010). Psychological methods 15:192-198.
610
Danzer SC (2012) Depression, stress, epilepsy and adult neurogenesis. Exp Neurol 233:22-32.
611
612
613
Danzer SC, Crooks KR, Lo DC, McNamara JO (2002) Increased expression of brain-derived
neurotrophic factor induces formation of basal dendrites and axonal branching in dentate granule
cells in hippocampal explant cultures. J Neurosci 22:9754-9763.
614
615
Danzer SC, McMullen NT, Rance NE (1998) Dendritic growth of arcuate neuroendocrine
neurons following orchidectomy in adult rats. J Comp Neurol 390:234-246.
616
617
618
Encinas JM, Michurina TV, Peunova N, Park JH, Tordo J, Peterson DA, Fishell G, Koulakov A,
Enikolopov G (2011) Division-coupled astrocytic differentiation and age-related depletion of
neural stem cells in the adult hippocampus. Cell stem cell 8:566-579.
619
620
Fox JG, Barthold SW, Davisson MT, Newcomer CE, Quimby FW, Smith AL (2007) The Mouse
in Biomedical Research: Normative Biology, Husbandry, and Models: Academic press.
621
622
Gaarskjaer FB, Laurberg S (1983) Ectopic granule cells of hilus fasciae dentatae projecting to
the ipsilateral regio inferior of the rat hippocampus. Brain Res 274:11-16.
623
624
Gray WP, Sundstrom LE (1998) Kainic acid increases the proliferation of granule cell
progenitors in the dentate gyrus of the adult rat. Brain Res 790:52-59.
625
626
627
Hama H, Kurokawa H, Kawano H, Ando R, Shimogori T, Noda H, Fukami K, Sakaue-Sawano
A, Miyawaki A (2011) Scale: a chemical approach for fluorescence imaging and reconstruction
of transparent mouse brain. Nat Neurosci 14:1481-1488.
628
629
Hattiangady B, Rao MS, Shetty AK (2004) Chronic temporal lobe epilepsy is associated with
severely declined dentate neurogenesis in the adult hippocampus. Neurobiol Dis 17:473-490.
630
631
632
Hattiangady B, Shetty AK (2010) Decreased neuronal differentiation of newly generated cells
underlies reduced hippocampal neurogenesis in chronic temporal lobe epilepsy. Hippocampus
20(1):97-112.
28
633
634
Hester MS, Danzer SC (2013) Accumulation of abnormal adult-generated hippocampal granule
cells predicts seizure frequency and severity. J Neurosci 33:8926-8936.
635
Hope K, Bhatia M (2011) Clonal interrogation of stem cells. Nature methods 8:S36-40.
636
637
638
Jessberger S, Zhao C, Toni N, Clemenson G, Li Y, Gage F (2007) Seizure-associated, aberrant
neurogenesis in adult rats characterized with retrovirus-mediated cell labeling. The Journal of
Neuroscience 27:9400-9407.
639
640
641
642
Jhaveri DJ, O'Keeffe I, Robinson GJ, Zhao QY, Zhang ZH, Nink V, Narayanan RK, Osborne
GW, Wray NR, Bartlett PF (2015) Purification of neural precursor cells reveals the presence of
distinct, stimulus-specific subpopulations of quiescent precursors in the adult mouse
hippocampus. J Neurosci. 35(21):8132-44.
643
644
Jinno S (2011) Topographic differences in adult neurogenesis in the mouse hippocampus: a
stereology-based study using endogenous markers. Hippocampus 21:467-480.
645
646
Kempermann G, Jessberger S, Steiner B, Kronenberg G (2004) Milestones of neuronal
development in the adult hippocampus. Trends Neurosci 27:447-452.
647
648
Kempermann G, Song H, Gage FH (2015) Neurogenesis in the Adult Hippocampus. Cold Spring
Harbor perspectives in biology 7:a018812.
649
650
651
Kheirbek MA, Hen R (2011) Dorsal vs ventral hippocampal neurogenesis: implications for
cognition and mood. Neuropsychopharmacology : official publication of the American College
of Neuropsychopharmacology 36:373-374.
652
653
Kron MM, Zhang H, Parent JM (2010) The developmental stage of dentate granule cells dictates
their contribution to seizure-induced plasticity. J Neurosci. 30(6):2051-9.
654
655
Lai K, Kaspar BK, Gage FH, Schaffer DV (2003) Sonic hedgehog regulates adult neural
progenitor proliferation in vitro and in vivo. Nat Neurosci 6:21-27.
656
Lange K (2003) Applied Probability. Springer Text in Statistics.
657
658
Ledergerber D, Fritschy JM, Kralic JE (2006) Impairment of dentate gyrus neuronal progenitor
cell differentiation in a mouse model of temporal lobe epilepsy. Exp Neurol 199:130-142.
29
659
660
661
Livet J, Weissman TA, Kang H, Draft RW, Lu J, Bennis RA, Sanes JR, Lichtman JW (2007)
Transgenic strategies for combinatorial expression of fluorescent proteins in the nervous system.
Nature 450:56-62.
662
663
Loi M, Koricka S, Lucassen PJ, Joels M (2014) Age- and sex-dependent effects of early life
stress on hippocampal neurogenesis. Frontiers in endocrinology 5:13.
664
665
Lu J, Fiala JC, Lichtman JW (2009) Semi-automated reconstruction of neural processes from
large numbers of fluorescence images. PLoS One 4:e5655.
666
667
668
McAuliffe JJ, Bronson SL, Hester MS, Murphy BL, Dahlquist-Topala R, Richards DA, Danzer
SC (2011) Altered patterning of dentate granule cell mossy fiber inputs onto CA3 pyramidal
cells in limbic epilepsy. Hippocampus 21:93-107.
669
670
Ming GL, Song H (2011) Adult neurogenesis in the mammalian brain: significant answers and
significant questions. Neuron 70:687-702.
671
672
673
Murphy BL, Hofacer RD, Faulkner CN, Loepke AW, Danzer SC (2012) Abnormalities of
granule cell dendritic structure are a prominent feature of the intrahippocampal kainic acid model
of epilepsy despite reduced postinjury neurogenesis. Epilepsia 53:908-921.
674
675
Murphy BL, Pun RY, Yin H, Faulkner CR, Loepke AW, Danzer SC (2011) Heterogeneous
integration of adult-generated granule cells into the epileptic brain. J Neurosci 31:105-117.
676
677
Myers CE, Bermudez-Hernandez K, Scharfman HE (2013) The influence of ectopic migration of
granule cells into the hilus on dentate gyrus-CA3 function. PLoS One 8:e68208.
678
679
680
Nakahara S, Tamura M, Matsuki N, Koyama R (2009) Neuronal hyperactivity sustains the basal
dendrites of immature dentate granule cells: time-lapse confocal analysis using hippocampal
slice cultures. Hippocampus 19:379-391.
681
682
683
Palma V, Lim DA, Dahmane N, Sanchez P, Brionne TC, Herzberg CD, Gitton Y, Carleton A,
Alvarez-Buylla A, Ruiz i Altaba A (2005) Sonic hedgehog controls stem cell behavior in the
postnatal and adult brain. Development 132:335-344.
684
685
Parent JM, Elliott RC, Pleasure SJ, Barbaro NM, Lowenstein DH (2006) Aberrant seizureinduced neurogenesis in experimental temporal lobe epilepsy. Ann Neurol 59:81-91.
30
686
687
Parent JM, Janumpalli S, McNamara JO, Lowenstein DH (1998) Increased dentate granule cell
neurogenesis following amygdala kindling in the adult rat. Neurosci Lett 247:9-12.
688
689
690
Parent JM, Yu TW, Leibowitz RT, Geschwind DH, Sloviter RS, Lowenstein DH (1997) Dentate
granule cell neurogenesis is increased by seizures and contributes to aberrant network
reorganization in the adult rat hippocampus. J Neurosci 17:3727-3738.
691
692
Paxinos G, Franklin KBJ (2001) The Mouse Brain in Stereotaxic Coordinates. San Diego:
Academic Press.
693
694
695
Ribak CE, Tran PH, Spigelman I, Okazaki MM, Nadler JV (2000) Status epilepticus-induced
hilar basal dendrites on rodent granule cells contribute to recurrent excitatory circuitry. The
Journal of comparative neurology 428:240-253.
696
697
Rolando C, Taylor V (2014) Neural stem cell of the hippocampus: development, physiology
regulation, and dysfunction in disease. Current topics in developmental biology 107:183-206.
698
699
700
Santos VR, de Castro OW, Pun RY, Hester MS, Murphy BL, Loepke AW, Garcia-Cairasco N,
Danzer SC (2011) Contributions of mature granule cells to structural plasticity in temporal lobe
epilepsy. Neuroscience 197:348-357.
701
702
Scharfman H, Goodman J, McCloskey D (2007) Ectopic granule cells of the rat dentate gyrus.
Developmental neuroscience 29:14-27.
703
704
705
706
Scharfman HE, Goodman JH, Sollas AL (2000) Granule-like neurons at the hilar/CA3 border
after status epilepticus and their synchrony with area CA3 pyramidal cells: functional
implications of seizure-induced neurogenesis. The Journal of neuroscience : the official journal
of the Society for Neuroscience 20:6144-6158.
707
708
709
Scharfman HE, Pierce JP (2012) New insights into the role of hilar ectopic granule cells in the
dentate gyrus based on quantitative anatomic analysis and three-dimensional reconstruction.
Epilepsia 53 Suppl 1:109-115.
710
711
Shapiro LA, Ribak CE, Jessberger S (2008) Structural changes for adult-born dentate granule
cells after status epilepticus. Epilepsia 49 Suppl 5:13-18.
712
713
Sierra A, Martin-Suarez S, Valcarcel-Martin R, Pascual-Brazo J, Aelvoet SA, Abiega O,
Deudero JJ, Brewster AL, Bernales I, Anderson AE, Baekelandt V, Maletic-Savatic M, Encinas
31
714
715
JM (2015) Neuronal hyperactivity accelerates depletion of neural stem cells and impairs
hippocampal neurogenesis. Cell stem cell 16:488-503.
716
717
718
Song J, Zhong C, Bonaguidi MA, Sun GJ, Hsu D, Gu Y, Meletis K, Huang ZJ, Ge S, Enikolopov
G, Deisseroth K, Luscher B, Christian KM, Ming GL, Song H (2012) Neuronal circuitry
mechanism regulating adult quiescent neural stem-cell fate decision. Nature 489(7414):150-4.
719
720
721
Spigelman I, Yan XX, Obenaus A, Lee EY, Wasterlain CG, Ribak CE (1998) Dentate granule
cells form novel basal dendrites in a rat model of temporal lobe epilepsy. Neuroscience 86:109120.
722
723
724
Steiner B, Kronenberg G, Jessberger S, Brandt MD, Reuter K, Kempermann G (2004)
Differential regulation of gliogenesis in the context of adult hippocampal neurogenesis in mice.
Glia. 46(1):41-52.
725
726
727
Suh H, Consiglio A, Ray J, Sawai T, D'Amour KA, Gage FH (2007) In vivo fate analysis reveals
the multipotent and self-renewal capacities of Sox2+ neural stem cells in the adult hippocampus.
Cell stem cell 1:515-528.
728
729
Sutula T, He XX, Cavazos J, Scott G (1988) Synaptic reorganization in the hippocampus
induced by abnormal functional activity. Science 239:1147-1150.
730
731
732
733
Teixeira CM, Kron MM, Masachs N, Zhang H, Lagace DC, Martinez A, Reillo I, Duan X, Bosch
C, Pujadas L, Brunso L, Song H, Eisch AJ, Borrell V, Howell BW, Parent JM, Soriano E (2012)
Cell-autonomous inactivation of the reelin pathway impairs adult neurogenesis in the
hippocampus. J Neurosci 32:12051-12065.
734
735
736
Turski WA, Cavalheiro EA, Schwarz M, Czuczwar SJ, Kleinrok Z, Turski L (1983) Limbic
seizures produced by pilocarpine in rats: behavioural, electroencephalographic and
neuropathological study. Behavioural brain research 9:315-335.
737
738
739
Walter C, Murphy BL, Pun RY, Spieles-Engemann AL, Danzer SC (2007) Pilocarpine-induced
seizures cause selective time-dependent changes to adult-generated hippocampal dentate granule
cells. J Neurosci 27:7541-7552.
740
741
Zhan RZ, Timofeeva O, Nadler JV (2010) High ratio of synaptic excitation to synaptic inhibition
in hilar ectopic granule cells of pilocarpine-treated rats. J Neurophysiol 104:3293-3304.
742
32
743
Legends
744
Figure 1: (A) Timeline depicting the experimental paradigm used. To induce fluorophore
745
expression, mice were injected with tamoxifen three times during postnatal week 7 and
746
subsequently underwent either pilocarpine or saline treatment on postnatal week 8. Mice were
747
sacrificed on postnatal week 16. (B) Example of a 3-dimensional reconstruction of the mouse
748
hippocampus. The Scale cleared 300 µm sections were imaged, aligned and reconstructed into a
749
3-dimensional reconstruction of the hippocampus with single cell resolution. (C.1) Brainbow
750
fluorophore expression was absent from animals not treated with tamoxifen. (C.2) A small
751
cohort of animals were sacrificed two days after the last tamoxifen injection (in week 7) and
752
analysis of their dentate gyri revealed that the tamoxifen treatment induced, on average, two
753
type-1 cells (indicated by white arrows) per 300 µm hippocampal section. Clonal clusters were
754
observed in both control (C.3) and pilocarpine (C.4) treated animals. (D) The number of cells per
755
cluster increased in pilocarpine treated animals. Scale bar for B (3D reconstruction) = 600 µm;
756
C.1 and C.2 = 250 µm; C.3 and C.4= 200 µm.
757
758
Figure 2: (A) Cells present in clonal clusters were classified based on morphology (see methods)
759
into either (A.1) type-1 progenitor cells, (A.2) type-2/3 progenitor cells, (A.3) immature granule
760
cells, (A.4) mature granule cells with (A.5) spiny apical dendrites or (A.6) astrocytes.
761
(B) Immunocharacterization of the different cell types. Type-1 cells were shown to express nestin
762
and GFAP; Type-2/3 cells expressed doublecortin (DCX); Immature DGCs expressed calretinin;
763
Mature DGCs express calbindin and astrocytes were shown to express GFAP. (C) The number of
764
clonal clusters/mouse hippocampus was significantly increased in female mice that underwent
765
status epilepticus (SE) relative to female controls. Female SE mice also had more clusters than
766
male SE mice. (D) Graph shows the composition of cell types in clonal clusters from control and
33
767
SE animals. There was a significant decrease in the number of type-1 cells and a trend (p=0.06)
768
towards an increase in the number of mature cells in mice exposed to status. (E) The percentage
769
of clusters containing either type-1 or type 2/3 progenitors was decreased in SE mice relative to
770
controls, while the percentage of fully differentiated clusters increased. *, p<0.05; **, p<0.01.
771
Scale bar for A.1 – A.3 and A.6 = 25 µm; A.4 and A.5 = 50 µm; B = 20 µm. RFP, red
772
fluorescent protein. YFP, yellow fluorescent protein.
773
774
Figure 3: Ectopic dentate granule cells are derived from a small number of clonal clusters.
775
Shown is an image of clonal cluster composed entirely of hilar ectopic dentate granule cells
776
(higher magnification image in purple inset). The graph shows quantification of all the clusters
777
from SE animals which contained ectopic cells. Additionally, for comparison, nine randomly
778
selected clusters containing no ectopic cells are shown. Orange bars show the total number of
779
cells in the cluster whereas the blue bars show the number of ectopic cells. Ectopic DGCs tended
780
to occur in clusters in which majority of the cells are ectopic. GCL= granule cell layer, H= hilus.
781
Scale bar for A =150 µm, Scale bar for inset of A= 40 µm.
782
783
Figure 4: Dentate granule cells with basal dendrites arise from diverse of clonal clusters. Shown
784
is a neuronal reconstruction of a granule cell (red) with basal dendrites (white arrows) within a
785
clonal cluster. The axon is denoted by the arrowhead. Adjacent cells in the cluster are shown in
786
blue. The graph shows quantification of all the clusters from SE animals which contained cells
787
with basal dendrites (blue bars) relative to total cluster size (orange bars). For comparison, a
788
subset of randomly selected clusters that contained only normal DGCs are shown. Scale bar = 50
789
µm.
34
790
Figure 5: Graphs show the distribution of mature granule cells (top) and type-1 cells (bottom)
791
along the dorsal-ventral axis of the hippocampus. Black dots depict the total number of cells at
792
each bregma level (top graph only), while blue dots depict the number of mature or type-1 cells,
793
respectively. Red triangles give the percentage of mature or type-1 cells at each level. No
794
relationship between mature cells and bregma level was evident, while higher numbers and
795
proportions of type-1 cells were present at more dorsal levels.
796
797
Figure 6: Summary of the key findings of the study. The first panel shows five, type-1
798
progenitor cells labeled with either RFP (red) or YFP (yellow), numbered from 1 to 5. Under
799
control conditions most of the type-1 cells remain quiescent (progenitors 1, 2, 4 and 5 remain
800
quiescent), however, a proportion of type-1 cells will enter the mitotic cycle to give rise to
801
differentiated cells (only progenitor 3 undergoes terminal differentiation). After epileptogenesis,
802
three key changes occur: (i) The number of clusters containing type-1 cells decreases in epileptic
803
animals relative to controls, and the number of clusters composed of differentiated DGCs and
804
astrocytes increases (progenitors 1, 2, 4 and 5 terminally differentiate); (ii) Progenitor cells either
805
produce all correctly located offspring, or ectopic offspring (progenitor 5 gives rise to a cluster
806
composed entirely of ectopic DGCs) and (iii) progenitors that produce correctly located offspring
807
occasionally produce cells with a basal dendrite, but mostly produce cells with normal dendrites
808
(progenitors 1 and 2 give rise to clusters which contain DGCs with basal dendrites and normal
809
DGCs).
35
Table 1
Data
Number of clonal
clusters per
hippocampus
Type of test
Two-way ANOVA
with treatment
and sex as factors
power
Treatment: 0.372
Treatment x Sex: 0.579
Two-way ANOVA
with treatment
and sex as factors
Treatment: 0.516
Treatment x Sex: 0.259
Equal Variance Test (BrownForsythe):Passed(P = 0.399)
Normality Test (Shapiro-Wilk):
Failed (P < 0.050)
Mann-Whitney
Rank Sum Test
25-75% confidence
intervals, control: 7.7637; SE: 0-3.5
Control vs SE, Type 2
Normality Test (Shapiro-Wilk):
Failed (P < 0.050)
Mann-Whitney
Rank Sum Test
25-75%, control: 0-21.6;
SE: 0-6.1
Control vs SE, Immature
cells
Normality Test (Shapiro-Wilk):
Failed (P < 0.050)
Mann-Whitney
Rank Sum Test
25-75%, control: 0-23.4;
SE: 0-4.5
Control vs SE, Mature
cells
Normality Test (Shapiro-Wilk):
Failed (P < 0.050)
Mann-Whitney
Rank Sum Test
25-75%, control: 26.983.6; SE: 75.0-93.8
Control vs SE,
Astrocytes
Normality Test (Shapiro-Wilk):
Failed (P < 0.050)
Mann-Whitney
Rank Sum Test
25-75%, control: 0-10.4;
SE: 3.2-7.0
Clusters with
progenitors
Self-renewing (2 type 1)
clusters
Normality Test (Shapiro-Wilk):
Passed (P = 0.526)
Normality Test (Shapiro-Wilk):
Failed (P < 0.050)
t-test
Power = 0.793
Mann-Whitney
Rank Sum Test
25-75%, control: 0-0.15;
SE: 0-0
Fully differentiated
clusters
Ectopic cells
Normality Test (Shapiro-Wilk):
Passed (P = 0.536)
Normality Test (Shapiro-Wilk):
Failed (P < 0.050)
t-test
Power = 0.790
Mann-Whitney
Rank Sum Test
25-75%, control: 0-0;
SE: 0-0
Cells with basal
dendrites
Normality Test (Shapiro-Wilk):
Failed (P < 0.050)
Mann-Whitney
Rank Sum Test
25-75%, control: 0-0;
SE: 0-0
Mean clone size
Control vs SE, Type 1
Data Structure
Normality Test (Shapiro-Wilk):
Passed (P = 0.561)
Equal Variance Test (BrownForsythe):Passed(P = 0.614)
Normality Test (Shapiro-Wilk):
Passed (P = 0.658)
1